Indonesia, as a tropical region and prone to frequent thunderstorms, is faced with potential risks to both human life and infrastructure. The analysis of a dataset containing observations of thunderstorms enabled us to identify variations in the vertical air temperature profiles between days experiencing thunderstorms and those characterized by no significant weather conditions (Nosig) within the layer between 1000 and 150 hPa. Examination of the relative humidity in the middle-troposphere during the thunderstorm days exhibited elevated moist layers in compared to the Nosig days across all the investigated regions. By employing stability indices that reflect the atmospheric conditions favorable for thunderstorm development, in conjunction with gridded reanalysis data and logistic regression methodologies, we ascertained that among the numerous convective stability parameters scrutinized, precipitable water (PW), K index (KI), and relative humidity in the middle troposphere (RH_Middle) emerged as the most effective parameters in distinguishing between environmental conditions conducive to thunderstorms and those devoid of significant weather phenomena. Assessment utilizing receiver operating characteristic curves illustrated that the optimal normalized thresholds for PW, KI, and RH_Middle were 0.67, 0.86, and 0.62, respectively.

Quantum computers have attracted much attention in recent years. This is because the development of the actual quantum machine is accelerating. Research on how to use quantum computers is active in the fields such as quantum chemistry and machine learning, where vast amounts of computation are required. However, in weather and climate simulations, less research has been done despite similar computational demands. In this study, a quantum computing algorithm is applied to a problem of the atmospheric science. The effectiveness of the proposed algorithm is evaluated using a quantum simulator. The results show that it can achieve the same simulations as a conventional algorithm designed for classical Ueno and Miura, Quantum Algorithm for a Stochastic Multicloud Model computers. More specifically, the stochastically 12 fluctuating behavior of a multi-cloud model was obtained using classical Monte Carlo method, and comparable results are also achieved by utilizing probabilistic outputs of computed quantum states. Our results show that quantum computers have a potential to be useful for the atmospheric and oceanic science, in which stochasticity is widely inherent.

 Local wind systems, including land and sea breezes, are significant factors influencing temperature distribution over the Kanto Plain. In this study, the diurnal variation patterns of these local wind systems on sunny summer days were classified into five categories based on the divergence field using dense surface observation data. Furthermore, the characteristics of temperature distribution and pressure fields, as well as recent changes in their frequency were examined.

 Each wind category was closely related to the pressure gradient around Japan, revealing differences in temperature distribution based on the wind pattern. In categories with a large northward pressure gradient, a North Pacific subtropical high south of Japan was strong, and the surface wind systems were dominated by southerly winds, causing sea breeze fronts to penetrate quickly. In these categories, notably high daytime temperatures were observed in the eastern Kanto region, where the easterly sea breeze was weaker than average. The frequency of these categories has increased in recent years, likely corresponding to the more frequent appearance of the pressure pattern where the subtropical high extends south of Japan.

Over the Northwest Pacific off Hokkaido, low clouds such as stratocumulus, stratus, and fog form frequently in summer. Despite the scientific and socioeconomic importance of these low clouds, our understanding of their reproducibility in weather prediction models under different synoptic circumstances is still lacking. This study assesses the ability of low-cloud representation and prediction in the Japan Meteorological Agency Meso-Scale Model (MSM) mainly in June and July 2020-2022, focusing particularly on synoptic fluctuations of near-surface temperature advection and low-cloud fraction. The low-cloud fraction predicted from the preceding day corresponds quite well with that in the analysis. However, the low-cloud fraction in the analysis is not highly correlated with satellite observations on daily timescales, while the mean low-cloud fraction agrees with the observations. The MSM sometimes simulates unrealistic fog within the near-surface stable layer associated with warm advection, resulting in a positive cloud fraction bias and a negative downward shortwave radiation bias. Under both warm and cold advection, the MSM may also underestimate the low-cloud fraction, although it still reproduces the horizontal distribution of low clouds rather consistently with satellite observations.

In this study, we conducted large-eddy simulations of turbulent flows and plume dispersion over idealized two-dimensional double steep hills. In the simulations, we investigated the distribution patterns of the mean plume concentrations, considering various distances between the hills and emission sources. Our objective was to provide information on the area of influence of local hilly terrains on plume dispersion from the viewpoint of accuracy, i.e., determining if the conventional Gaussian plume model can accurately predict plume concentrations. The result showed that the clockwise circulation was dominant in the area between the windward and leeward hills (valley) when the valley width was less than 10 times the hill height (H). This circulation makes the flow close to the stack remain in the valley, resulting in the higher concentrations in the valley than in wider-valley (> 10H valley) cases. The effect of the leeward hill on the flow field was negligible when the valley width was greater than 10H. In the area beyond 20H from the crest of the windward hill, estimated plume spreads for all cases were similar, indicating that the area of influence of the hills was approximately 20H.

The Weather Research and Forecasting model version 4.3.3 with an eddy-diffusivity mass-flux (EDMF) scheme, which is a blend of the Mellor–Yamada–Nakanishi–Niino scheme including a partial condensation scheme and a mass-flux scheme, is used to forecast a quasi-stationary band-shaped precipitation event that occurred over the Hokuriku region on 12-13 July 2023. However, the original model does not satisfactorily reproduce the heavy precipitation over the region at a horizontal resolution of 5 km. Our companion papers suggested that improving the parameterization of the buoyancy production of turbulent kinetic energy (TKE) is necessary for such a coarse-resolution model to accurately simulate the transition from shallow to deep convection that triggers heavy precipitation. In this study, its parameterization in the EDMF scheme is modified to divide it into two contributions from the partial condensation and mass-flux schemes. The results show that although the modified model predicts the band-shaped distribution of precipitation with a westward bias, it simulates the timing and intensity of heavy precipitation reasonably well. This study presents the modified parameterization of the TKE buoyancy production and demonstrates that a modification of its parameterization can improve the prediction of heavy precipitation.